🎙️ Gemini Live Web Socket Server

A Python ASGI Socket.IO server that proxies to Google’s Gemini API over WebSockets

Bridging real-time clients with powerful AI services, complete with JWT authentication, CORS support, and metrics!

🚀 Features

• 🔄 Real-time bi-directional messaging via Socket.IO
• 🌐 ASGI server built on FastAPI + Uvicorn
• 🛡️ JWT Authentication (or pluggable auth middleware)
• 🔧 Tool Presets for translate_text, summarize_text, generate_code calls
• 🔄 WebSocket Bridge to Gemini (models/gemini-2.0-flash-exp)
• 📊 Optional Metrics (active connections, processed messages, errors)
• 🛬 Graceful Startup & Shutdown
• 🔧 Configurable : CORS, retry/backoff, hooks, debug logging

📦 Installation

1. Clone the repo

   git clone https://github.com/omanandswami2005/gemini_live_ws_server_pipy.git
   cd gemini_live_ws_server_pipy
   

2. Create & activate a virtual environment

   python -m venv myenv
   source myenv/bin/activate   # macOS/Linux
   myenv\Scripts\activate    # Windows
   

3. Install dependencies

   pip install .             # published on PyPI
   

Tip: you can also install directly from PyPI:

pip install --index-url https://pypi.org/simple/gemini_live_ws_server

⚙️ Configuration

Create a .env file or pass flags to your script:

GOOGLE_API_KEY=your_google_api_key_here
JWT_SECRET=super_secret_key
PORT=8080
CORS_ORIGINS=http://localhost:3000,http://yourdomain.com
DEBUG=true

🔑 Environment Variables :

• GOOGLE_API_KEY: 🔗 Your Google Cloud API key with Generative API access
• JWT_SECRET: 🛡️ Secret for signing/verifying Socket.IO tokens
• PORT: 🖥️ Server listening port (default: 8080)
• CORS_ORIGINS: 🌐 Comma-separated origins for CORS
• DEBUG: 🐞 true or false for verbose logging

🎬 Usage

python -m gemini_live_ws_server.main_server

OR using Uvicorn directly:

uvicorn gemini_live_ws_server.main_server:server --host 127.0.0.1 --port 8080 --reload

Once running, clients can connect to ws://your_host:PORT/socket.io/ with a valid JWT:

import io from 'socket.io-client';

const socket = io('http://localhost:8080', {
  auth: { token: '<YOUR_JWT_TOKEN>' }
});

socket.on('ready', () => console.log('🟢 Connected!'));
socket.on('message', msg => console.log('📨 AI:', msg));

💡 Examples

Translate Text

socket.emit('message', {
  toolCall: {
    functionName: 'translate_text',
    arguments: { text: 'Hello, world!', targetLanguage: 'es' }
  }
});

Summarize Text

socket.emit('message', {
  toolCall: {
    functionName: 'summarize_text',
    arguments: { text: longParagraph, maxLength: 50 }
  }
});

📝 Contributing

1. ⭐ Star this repo
2. 🔀 Fork it
3. ✨ Create a feature branch (git checkout -b feature/new-thing)
4. 🛠️ Make changes & commit (git commit -m 'feat: add awesome feature')
5. 🔄 Push to the branch (git push origin feature/new-thing)
6. 📥 Open a Pull Request

We welcome bug reports, feature requests, and code contributions! ❤️

📜 License

This project is licensed under the Apache 2.0

Made with ❤️ by omiii 😅